In March 2020, most of the world implemented non-pharmaceutical interventions (NPIs) including the use of face masks, social distancing, school closures, and teleworking to prevent the spread of Covid-19. We don’t yet know exactly how well these interventions worked to slow the Covid-19 pandemic. But we do know now that they have helped to slow the spread of influenza.
Ideally, the effectiveness of an intervention (e.g. wearing a face mask), is measured by looking at the outcomes one is trying to affect (e.g. the number of cases of Covid-19). With seasonal diseases like flu, the ability to compare the current year to previous years makes this easy. But with Covid-19, we have no history for comparison. It’s also difficult to assess the effectiveness of interventions on Covid-19 by other means due to variable test availability, lack of study design, and changing criteria to receive a test. So some deniers have argued that NPIs have had little epidemiological impact.
But since influenza, like Covid-19, is a respiratory disease, it is reasonable to suppose interventions targeting Covid-19 should also affect the spread of flu. And if we see Covid-19 NPIs working on the flu, we should infer they are working to some degree on Covid-19, too. Now, a new study in the CDC’s Morbidity and Mortality Weekly Report (MMWR) looks at the relationship between national responses to Covid-19 and the decline in cases of influenza in the US and three Southern Hemisphere countries
In the US, there was a clear difference in the trajectory of this year’s flu season, timed to the arrival of Covid-19.
First, the 2020 summer flu all but disappeared. Although not widely known, people can ‘catch the flu’ anytime of year, not just during the winter. Influenza transmission intensity is measured with the positivity rate, the fraction of respiratory infections that test positive for influenza. During a typical US summer, influenza positivity tends to be around 1% or 2%. In 2017, it was 2.36%. In 2018 it was 1.04%. In 2019 it was 2.35%. And in the first summer with Covid-19, it was reduced five to ten fold to 0.20%.
Second, the arrival of Covid-19 in the U.S. (marked by the President’s declaration of a state of national emergency) coincided with the decline of the usual winter influenza season. By comparing the speed at which the 2020 flu seasons “ended” compared with other years, the authors could get an idea about how NPIs cut off flu transmission. Whereas most flu seasons tail off gradually, the 2020 season stopped precipitously. If we count the “end” of the influenza season as the time that the positivity rate drops to it’s normal inter-seasonal level around 2%, it appears that the 2020 flu season ended almost two months early.
Finally, the authors looked at influenza in three countries — Australia, Chile, and South Africa — in the Southern Hemisphere. This is important because winter in the Southern Hemisphere corresponds to summer in the Northern Hemisphere. Just as the U.S. influenza season is winding down, influenza in the Southern Hemisphere should be heating up. But that’s not what happened in 2020. This year, influenza has been almost undetectable in the Southern Hemisphere with Australia, Chile, and South Africa reporting a total of only 51 positive specimens out of 83,307 samples tested for an unheard of positivity rate of 0.06%. For comparison, the average positivity during 2017-2019 in these countries was 13.7% meaning that cases of influenza in the Southern Hemisphere have dropped by more than 99%.
The upshot of these analyses is that the NPIs adopted to mitigate the spread of Covid-19 are almost certainly also responsible for drastic reductions in the transmission of influenza. And since they are reducing influenza transmission, they are probably reducing Covid-19 transmission as well. NPIs work.
The study also highlights data that will be of interest to people following the Covid-19 pandemic who might like to compare it with flu. The study used two different data sets. Data from the US are from hundreds of diagnostic laboratories that report to either the WHO Collaborating Laboratories System or the National Respiratory and Enteric Virus Surveillance System. Data from Australia, Chile, and South Africa are from the Global Influenza Surveillance and Response System. All data used in the study are open access, meaning they are in the public domain, provided by public agencies for scientists and laypeople alike. Data for the US are available to download from the CDC’s FluView platform. Data from Australia, Chile, and South Africa can be obtained from the World Health Organization’s FluNet system. Both websites provide interactive tools allowing anyone to select and explore data they find interesting. Just as amateur astronomers have discovered exoplanets, I bet there are epidemiological patterns waiting to be discovered by anyone with the interest, imagination and diligence to look for them.